IMPC- Understanding somatic MSI+ colorectal cancers using conditional coding and non-coding mutations in Mlh1.

Colorectal cancer affects 1 in 20 people in the western world. Mutations that cause colon cancer can be inherited or, more often, acquired during a patient's lifetime. One commonly mutated gene is MLH1 which is important for repairing mistakes or "mismatches" in DNA sequences. Mutation in the coding sequence of the MLH1 gene can be inherited, causing the famililal cancer syndrome called Lynch. However they are very rarely acquired in somatic tissues. Instead, MLH1 is normally repressed by chemical marks on the DNA (called DNA methylation or epimutation). Both inherited and acquired mutations in MLH1 disrupt DNA repair and lead to high levels of mutations in the DNA which are key to the development of cancer. However, the cancers that arise from the two types of mutations have important differences in their prognosis and response to treatment, so we need to investigate them using separate and appropriate systems.
Mouse models of cancer provide powerful ways of studying the underlying biology of cancer development because, unlike isolated cell line systems, they allow us to study tumour cells in an appropriate context or environment which can dramatically change the way they behave. Crucially they can also be used for the testing of potential treatments. Currently there are models carrying Mlh1 mutations that are representative of Lynch syndrome but none that "acquire" Mlh1 repression together with an appropriate combination of other mutations in colon cancer genes. In this project we will aim to do this using mice carrying a "conditional" mutation in Mlh1 that can be activated in specific tissues at specific times. We will combine the conditional mutation with a common cancer causing mutation in the Braf oncogene and activate them in cells within the intestine in adult mice. We will also combine a promoter mutation, similar to an epimutation, that partially represses Mlh1 expression with the same intestinal Braf mutation. We will then characterise the type, severity and biology of the cancers arising in mice carrying these mutation combinations. We propose that one or both of these models will provide a suitable system for testing existing and new drugs, alone or in combinations, specifically designed to treat the highly mutated cancers caused by acquired mutations in Mlh1.

Microsatellite-unstable (MSI+) tumours are commonly found in the colon where they comprise ~15% of all cancers. MSI+ cancers can occur as a result of the familial Lynch syndrome which is caused by germline mutations in components of the mismatch repair pathway such as MSH2 and MLH1. MSI can also arise in sporadic cancers where, in the majority of cases, MLH1 is silenced by promoter methylation. Colorectal tumours arising in Lynch syndrome patients usually develop via the classical adenoma to carcinoma pathway. In contrast, sporadic MSI+ cancers develop from sessile serrated adenomas. This pathway is characterised by the presence of BrafV600E mutations and has a lower frequency of APC mutations than the classical pathway.
Sporadic MSI+ cancers with BrafV600E mutations and promoter silencing of MLH1 form a significant subgroup of colorectal cancer with a good prognosis but a differential response to common chemotherapies. A greater understanding of this group of cancers is essential to develop subtype specific or targeted treatments.
Germline Mlh1 mouse knockouts with and without Apc mutations have been studied and proved representative of Lynch syndrome but to date, no model of somatic Mlh1 coding mutation combined with BrafV600E mutation exists. In this proposal we aim to generate and characterise such a model using the IMPC Mlh1tm1c strain and compare it with an existing Mlh1 promoter mutation created in our laboratory. We will use these systems to investigate the initiation and progression of sporadic MSI+ tumours. Our long term goal will be to use either or both models to test existing and novel therapies, including immunological and epigenetic based approaches, in this distinct class of colorectal cancer.